INDEX

Explanations

words related to investigative or exploratory activities

oai_token-act-pair · gpt-3.5-turbo

terms related to investigative and exploratory actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fol

-0.79

 Leilan

-0.78

frey

-0.77

enberg

-0.75

ovan

-0.73

ned

-0.71

nings

-0.70

 Machina

-0.69

ahar

-0.67

imil

-0.67

POSITIVE LOGITS

atory

0.95

ItemTracker

0.79

 sidx

0.75

 sqor

0.75

allery

0.72

orically

0.72

 bowel

0.68

 easing

0.65

 pronoun

0.65

 enqu

0.65

Activations Density 0.032%

words related to investigative or exploratory activities

terms related to investigative and exploratory actions

No Comments

No Known Activations